Update app.py

app.py

@@ -8,8 +8,9 @@ model = AutoModelForCausalLM.from_pretrained(
     trust_remote_code=True,
     device_map="auto",
     low_cpu_mem_usage=True,
+    device_map = {"": 0}
 )
-tokenizer = AutoTokenizer.from_pretrained("Trelis/Llama-2-7b-chat-hf-sharded-bf16")
+tokenizer = AutoTokenizer.from_pretrained("Trelis/Llama-2-7b-chat-hf-sharded-bf16", trust_remote_code=True,)
 
 
 def generate_text(input_text):
@@ -19,7 +20,6 @@ def generate_text(input_text):
     output = model.generate(
         input_ids,
         attention_mask=attention_mask,
-        max_length=200,
          max_new_tokens=400, top_p = 0.7, top_k = 50,
         do_sample=True,
         num_return_sequences=1,